Real-time scheduling algorithm for safety-critical systems on faulty multicore environments

Abstract

An algorithm (called FTM) for scheduling of real-time sporadic tasks on a multicore platform is proposed. Each task has a deadline by which it must complete its non-erroneous execution. The FTM algorithm executes backups in order to recover from errors caused by non-permanent and permanent hardware faults. The worst-case schedulability analysis of FTM algorithm is presented considering an applicationlevel error model, which is independent of the stochastic behavior of the underlying hardware-level fault model. Then, the stochastic behavior of hardware-level fault model is plugged in to the analysis to derive the probability of meeting all the deadlines. Such probabilistic guarantee is the level of assurance (i.e., reliability) regarding the correct functional and timing behaviors of the system. One of the salient features of FTM algorithm is that it executes some backups in active redundancy to exploit the parallel multicore architecture while other backups passively to avoid unnecessary execution of too many active backups. This paper also proposes a scheme to determine for each task the number of backups that should run in active redundancy in order to increase the probability of meeting all the deadlines. The effectiveness of the proposed approach is demonstrated using an example application.

Fault-Tolerant Online Scheduling Algorithms for CubeSats

“…The data used in our experiment framework are based on real CubeSat data provided by the Auckland Program for Space Systems (APSS) 6 . These data were gathered by functionality and generalised in order to generate more data for our simulations.…”

“…CubeSats operate in harsh space environment and are exposed to charged particles and radiations, which cause transient effects, such as single event upsets, and long-term effects like total ionising dose [5]. Fault rate can be up to 10 -2 fault per second [6]. Besides natural faults, there are also human faults due to for instance a bug in software or wrong commands [7].…”

Evaluation of Fault Tolerant Online Scheduling Algorithms for CubeSats

“…For simulations with fault injection, we take into account that the worst estimated fault rate in the real space environment is 10 −5 fault/ms [18]. Therefore, we inject faults at the level of task copies with fault rate for each processor between 1 • 10 −5 and 1 • 10 −3 fault/ms in order to assess algorithm performances not only using the real fault rate but also its higher values.…”

Timing Predictability in Future Multi-Core Avionics Systems

“…Pathan also presents a global scheduling algorithm (FTM) for real-time sporadic tasks on multi-core platforms [72], but with no focus on mixedcriticality (this is left as future work). The algorithm considers a combination of active and passive redundancy, where active backups are executed in parallel on a different core from the primary task and passive backups are executed in sequence.…”

“…In the integration phase where all tasks on all cores are brought together the final WCET (or WCRT) can be determined and schedulability tests can be performed. The works considering schedulability [3,36,44,71,72] all assume that WCET estimates have been derived and are ready to be used. The response times estimated in these works have to be interference-aware as the fault tolerance method used is re-execution of a task, which may introduce additional memory requests that interfere with or are interfered with by tasks on other cores resulting in a variability of WCRT.…”